We’ve jiggled our sign-up buttons around and re-wordified them. We’ve also modified the flow of signing up for Via services (hopefully for the better), and corrected some validation message silliness.
Whilst he was at it, and because he has sleepless nights worrying about this sort of thing, our resident wordsmith (slacker) tweaked the word case of a few buttons, and twiddled with the sign-up emails generated by the system.
Continuing the theme of trying to be a little more helpful in the web application itself, we have implemented a session time-out warning. If you leave your Via session idle for any lengthy period of time, a helpful little doo-dad will appear on-screen telling you as much, with a countdown to when your session will expire.
We have a new pricing structure, with upgraded account limits. See our pricing page for more details.
Users weren’t getting a complete picture when listing all assigned databases in Via. That picture is now complete.
System admins (OK, us) can now delete organisations properly, and all users can display organisation data properly. We can also flag user passwords for reset — rest assured that security remains paramount: we can flag your password for a reset, but we can’t access the password itself, nor would we want to.
We haven’t left out organisation super-users! You now have a more performant super-user dashboard, and that’s, er, super. Super-users can also remove database access for given users properly, and we’ve incorporated some improved error-handling when rendering both organisation details and lists of users.
When super-user and system administrators (us again) do certain things like editing, deleting or resetting user accounts, these actions are now logged. Don’t worry, we won’t tell anyone, it just makes the support process a little easier.
… In fact, we’ve improved error logging and handling across the board, especially in terms of dealing with malformed rich text and error-prone incoming data migrations.
API & metadata
Oh, so many API enhancements! You can now reset passwords, see who can access a given database, and even delete users properly. What is more, you can now PUT arrays of documents to the API, which gives the collections interface a rather splendid performance boost. In addition, we’ve fixed a couple of issues with the “get other user details” and database activity APIs.
Not content with API updates, we’ve also enhanced how we work with metadata in Via. We have added a “template” property when getting database stats; and, more interestingly, developers can now set and retrieve additional unstructured metadata at the database details level. All sorts of possibilities for applications built using Via as the back-end!
There are some shady characters we no longer allow in database names. They were causing havoc frankly, so database names have had a colectomy.
We performed some housekeeping around drivers and low-level stuff: a fair bit of our code is now ES6-compliant (that sounds fancy), our MongoDB drivers are up-to-date and we also gave our email pixies a bit of a pep talk.
Users can add more attachments to documents that already have attachments without making a mess.
Not only that, but documents with complex objects stored in fields are also properly supported (nested JSON ahoy). Crikey!
When importing data, we were seeing the occasional issue with
__modified date fields (specifically, when adding new documents). This bug has been squashed.
In the low-level data import code, we have shifted our bits so that user account registration can happen without having to send automated registration emails. This means that organisations have more control over their data migration and application provisioning projects: we let you manage the message if you want to.